Conventional stereoscopic 3D displays create the illusion of depth based on binocular disparities, rendering 3D scenes from a single pair of 2D perspective images at a fixed distance to the viewer. Therefore conventional stereoscopic displays force an unnatural decoupling of the accommodation and convergence cues, which may contribute to various visual artifacts in stereoscopic displays, such as distortion in perceived depth, diplopic vision, visual discomfort, and fatigue. Many approaches have been proposed which may overcome the drawbacks of conventional stereoscopic displays, including volumetric displays, holographic displays, and multi-focal-plane displays. However, a need exists to develop an optical see-through stereoscopic display which solves the fundamental accommodation-convergence problems and also renders large volumes of continuous 3D scene at high image quality and flickering-free speed.
A stereoscopic display with addressable focus cues is one of the most promising approaches to solving the fundamental accommodation-convergence problems of providing the capability of rendering correct or near-correct focus cues for virtual 3D objects. Unlike traditional stereoscopic displays, stereoscopic displays with addressable focus cues enable the ability to either dynamically vary the focal distance of the virtual display through an active optical element in accordance to the viewer's region of interest, known as vari-focal display mode, or present multiple focal planes at a flickering-free speed through with no need to track a viewer's region of interest, known as multi-focal display mode. For instance, multi-focal-plane displays present perspective 2D images at multiple carefully placed, discrete focal distances along the visual axis. These discrete focal planes sample a 3D scene volume into multiple zones, and objects within a zone are rendered by the corresponding pair of adjacent focal planes, as shown in FIG. 1. Therefore multi-focal-plane displays are able to render correct or near-correct focus cues for virtual objects at different depths. Contrary to multi-viewpoint displays such as holographic displays and volumetric displays, the multi-focal-plane display is a fixed-viewpoint display. By restricting the viewing position, multi-focal-plane display systems only need to display a small number of viewpoints. Also multi-focal-point displays can preserve disparity, occlusion and perspective in conventional 2D displays as well as rendering viewing-dependent lighting effects such as specular reflection and shading. In practice, the implementations of multi-focal-plane displays can be categorized into two categories: spatially multiplexed or temporally multiplexed. In a spatial-multiplexed system, multi-focal capability is achieved by stacking multiple 2D displays. In the alternative and more elegant, time-multiplexed system, the focal distances of images from a single 2D display are fast switched by an active optical element in synchronization with the frame rendering of multiple focal planes. In general, a multi-focal plane display can be readily adapted for use in a vari-focal mode without much change to the optics layout.
In addition, progress has recently been made in the field of stereoscopic displays that can be head-mounted and that have addressable focal planes for improved depth perception but require substantially less computational power than existing methods, as reflected in commonly owned U.S. Patent Application Publication. No. 2011/0075257, the contents of which are incorporated herein by reference. However, a need still exists for optical imaging systems which can provide enhanced imaging performance in stereoscopic displays with addressable focus cues.